[LINK] SBS: The Age of Big Data

[Bernard Robertson-Dunn]

On 6/01/2014 8:22 PM, stephen at melbpc.org.au wrote:

> So if you visit ANY McDonalds carrying your Android phone, with free WiFi
> turned on, then both McDonalds and Google will know. It's simplistc stats
> to know FOR SURE who visits, and when and what they buy. What idiot could
> not devise sure-fire business sales-plans from this absolutely hard data?

It's not necessarily hard data. There is an assumption that a person is 
linked with a particular phone.

> There is little especially AI/neural-networkish about absolute hard-data
> regarding customer behaviour. The data is simply times, dates, purchases.

The "hard data" is device, times data and purchases.

You do not know a) who is carrying the phone and b) who owns/eats the 
good or service (to generalise the scenario), once purchased.

A wife may have borrowed her husband's phone and she may have bought a 
meal for her daughter.

The assumptions underlying the data may often be correct but there is 
always the potential for it to be incorrect. This may not be very 
important when the data are treated statistically, but the spooks (or 
the suspect) could be in trouble if they rely on a single instance of a 
device being carried by a particular individual being in a particular 
place at a particular time. Even devices designed to track a specific 
individual can be error prone (see the last episode of The Blacklist) 
and phones and tablets are not inserted subcutaneously.

> And it's certainly a no-brainer.
>
> And, it's coming to every McDonalds near you.
>
> There's absolutely nothing naive about behavioral hard-big-data. And zero
> guesswork & little chance of being wrong.

See above.

> This data is just fact and drop dead easy to analyse. Analysed for targeting sales, and advertising plans.

I'm not sure that it actually adds much to sales data when the customer 
uses a credit/debit card.

> The only issue is, how *granulated* personal bigdata is allowed to become.

Don't agree. That's one issue, the other is accuracy in relationship 
between the device being monitored and the person of interest.

Data do not always reflect reality. The only question is, by how much 
and is it relevant.

-- 

Regards
brd

Bernard Robertson-Dunn
Sydney Australia
email: brd at iimetro.com.au
web:   www.drbrd.com
web:   www.problemsfirst.com
Blog:  www.problemsfirst.com/blog